Science of the Total Environment 912 (2024) 168825
Available online 27 November 2023
0048-9697/© 2023 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
Effects of agriculture on river biota differ between crop types and organism groups
Christian Schürings
a,*, Jochem Kail
a, Willem Kaijser
a, Daniel Hering
a,baDepartment of Aquatic Ecology, Faculty of Biology, University of Duisburg-Essen, Universit¨atsstrasse 5, D-45141 Essen, Germany
bCentre for Water and Environmental Research, University of Duisburg-Essen, Universit¨atsstrasse 5, D-45141 Essen, Germany
H I G H L I G H T S G R A P H I C A L A B S T R A C T
•Effects of catchment agriculture exceed urban effects on river biota in magnitude.
•Discrimination between crop types re- veals clear differences in biota response.
•Macroinvertebrates are most strongly impaired by pesticide intensive crops.
•Diatoms respond most strongly to the fertilization intensity of crop types.
•Macrophyte response is less clear and likely depending on river hydromorphology.
A R T I C L E I N F O Editor: Sergi Sabater Keywords:
Agriculture Biodiversity Catchment Freshwater Nutrients Pesticides Urban land use Watershed
A B S T R A C T
While the general effects of agricultural land use on riverine biota are well documented, the differential effects of specific crop types on different riverine organism groups, remain largely unexplored. Here we used recently published land use data distinguishing between specific crop types and a Germany-wide dataset of 7748 sites on the ecological status of macroinvertebrates, macrophytes and diatoms and applied generalized linear mixed models to unravel the associations between land use types, crop types, and the ecological status. For all organism groups, associations of specific crop types with biota were stronger than those of urban land use. For macro- invertebrates and macrophytes, strong negative associations were found for pesticide intensive permanent crops, while intensively fertilized crops (maize, intensive cereals) affected diatoms most. These differential associations highlight the importance of distinguishing between crop types and organism groups and the urgency to buffer rivers against agricultural stressors at the catchment scales and to expand sustainably managed agriculture.
1. Introduction
Agriculture is constantly intensifying to fulfill the needs of the world’s growing population (Foley et al., 2005), a process that is
accompanied by massive biodiversity decline. Besides terrestrial mam- mals, birds (Joppa et al., 2016) and flying insects (Hahn et al., 2015), the biota of rivers draining agricultural areas is also strongly impaired (Schürings et al., 2022).
* Corresponding author.
E-mail address: [email protected] (C. Schürings).
Contents lists available at ScienceDirect
Science of the Total Environment
journal homepage: www.elsevier.com/locate/scitotenv
https://doi.org/10.1016/j.scitotenv.2023.168825
Received 17 August 2023; Received in revised form 26 October 2023; Accepted 22 November 2023
Agriculture is frequently considered the most important driver for the deterioration of river biota (e.g. Wolfram et al., 2021). Calls for reducing agricultural impacts on rivers are manifold, particularly in parts of Europe where the majority of rivers do not meet the quality targets set by legislation; e.g., in Germany about 90 % of the river water bodies fail to meet the target of ‘good ecological status according to the EU Water-Framework Directive’ (UBA, 2022). On the global scale, the Convention on Biological Diversity adopted the need of pollution reduction and agricultural transition in targets 7 and 10 of the Kunming- Montreal global biodiversity framework (CBD, 2022). The imple- mentation of targeted environmental protection measures, however, is restricted by complex and contradictory legislation, land availability, and also because large-scale studies on stressors related to specific crop types and how these agricultural stressors affect different organism groups are rare, which would allow to draw more general conclusions for larger areas.
The first link in this cause-effect chain has already been investigated, but mainly in small-scale experimental studies. They found clear dif- ferences in stress on freshwaters arising from different crop types. While extensive grassland exerts less stress on river biota (Blake et al., 2012), corn farming close to rivers causes high nutrient and fine sediment influx (Secchi et al., 2011; Jwaideh et al., 2022) and permanent crops like orchards, vineyards, and vegetables are known for high pesticide treatment and runoff, particularly consisting of fungicides and in- secticides (Dachbrodt-Saaydeh et al., 2021). More specifically, pesti- cides cover a range of substances with different modes of action, including insecticides that may directly eradicate macroinvertebrates (Schulz and Liess, 1999), herbicides that may affect the growth of algae (Lorente et al., 2015) and higher plants (Ribeiro et al., 2019) once flushed into rivers and subsequently the feeding conditions of macro- invertebrates, and finally fungicides that may modify leaf litter break- down in streams and thus the food availability for invertebrates (Artigas et al., 2012).
The second link in this cause-effect chain has also already been investigated in large-scale studies, and there is clear empirical evidence that specific stressors which are potentially caused by agriculture affect river biota and that these effects differ between organism groups. Fine sediments (Davis et al., 2022) mainly affect macroinvertebrates (Davis et al., 2022) and Liess et al. (2021) identified pesticides applied on arable land as the most dominant stressor for vulnerable aquatic insects in lowland streams. In contrast, nutrients are more strongly associated with macrophytes (aquatic plants) and diatoms (O’Hare et al., 2018).
River morphology (Kaijser et al., 2022) and river management (Bączyk et al., 2018) are crucial variables for the occurrence of macrophytes, whereas morphological alterations directly impair macroinvertebrates (Urbaniˇc, 2014).
However, typical stressors resulting from agriculture can also origi- nate from other land use types, and contradictory results have been reported about whether agricultural or urban land use exerts higher stress on river biota. Streams draining urban areas are also frequently impacted by both hydromorphological alterations and nutrient pollution (Weitere et al., 2021), and, in some cases, also by pesticide pollution (Gerecke et al., 2002). As additional stressors are common, e.g., micropollutants and flash floods caused by impervious land cover (e.g.
Zhou et al., 2022), urban areas are frequently expected to exert higher stress on freshwater biota compared to agricultural land use (e.g.
B¨ohmer et al., 2004). In contrast, recent studies suggest agriculture, in particular cropland, as a main driver of freshwater biota deterioration (Gerecke et al., 2002; Neumann et al., 2002; Stehle and Schulz, 2015).
Most of these large-scale empirical studies only distinguished between rather broad land use categories and mainly used gross agricultural land use types, often only distinguishing between crop- and grassland (e.g.
Gieswein et al., 2017; Davis et al., 2022). Differential effects of specific crop types on river biota have rarely been considered in large-scale empirical studies yet (Wasson et al., 2010), as high-resolution land use data distinguishing specific crop types became available only recently
(e.g. Blickensdorfer et al., 2022). ¨
Against this background, a large dataset of 7748 sampling sites for macroinvertebrates, 2905 sites for macrophytes and 3402 sites for di- atoms from rivers across Germany was used to test if cropland exerts higher stress on river biota compared to other land use types (especially urban land use, but also grassland and forests), and to investigate the specific associations between different crop types and individual or- ganism groups (macroinvertebrates, macrophytes, diatoms). We hy- pothesized that (1) urban land use shows weaker associations with riverine biota compared to cropland and larger associations compared to grassland, while forests have positive associations. (2a) The associations between different crop types and the organism groups strongly differ due to cultivation practices such as crop-specific pesticide and nutrient application rates. (2b) Invertebrates are more sensitive to pesticide- intensive crops, while crops that are intensively fertilized are more strongly affecting macrophytes and diatoms.
2. Methods 2.1. Biological data
Data on biological samples (n =7748 for macroinvertebrates, n = 2905 for macrophytes, n =3402 for diatoms) taken between 2010 and 2019 were acquired from all German federal states, except the Saarland.
The biological samples were taken with standardized methods used for the ecological status assessment according to the EU Water Framework Directive. Macroinvertebrates were sampled with a multi-habitat sam- pling method (Haase et al., 2004); we calculated the river-type specific multimetric invertebrate index (MMI) with the resulting species-level taxa lists using the online tool PERLODES (https://www.gewaesser-be wertung-berechnung.de/index.php/perlodes-online.html). The MMI is a combination of biodiversity-related indices and was designed to reflect the impact of various stressors like hydromorphological degradation, altered hydrology, and impacts of land use (B¨ohmer et al., 2004; Hering et al., 2006). The German Fauna Index is the main metric included in the MMI, which mainly assesses the presence of river-type specific indicator species. Depending on the river-type, usually two to four additional core metrics are included in the MMI like the share of Ephemeroptera, Ple- coptera and Trichoptera (%EPT), the share of rheophilic species, or the number of Trichoptera species. Macrophytes and diatoms were sampled following Schaumburg et al. (2012) and the species-level taxa lists were processed using the online tool PHYLIB (https://www.gewaesser-bewer tung-berechnung.de/index.php/phylib-online.html) to calculate multi- metric indices for macrophytes and diatoms. The multimetric index for macrophytes is mainly based on the quantity of river-type specific in- dicator species, while the MMI for diatoms is based on both river-type specific indicator species and additional metrics mainly regarding nu- trients, salinization, or acidification. We used a standardized version of all three multimetric indices ranging from 0 (worst) to 1 (best) to enable later comparison. Metric selection and thresholds between quality classes differ between the 30 different German river-types, which ac- count for differences in ecoregion, altitude, geology, and river size (Pottgiesser and Sommerh¨auser, 2008). If sites have been sampled multiple times, we choose the sampling date closest to 2018 (one year after the land use data were recorded; see next paragraph).
2.2. Catchment land use
For each sampling site, we quantified catchment land use. Upstream catchments (watersheds) were automatically delineated with ESRI ArcView 3.3 based on a digital elevation model (DEM, 10 m resolution) and visually checked. We used agricultural land use data for the year 2017 provided by Blickensd¨orfer et al. (2022) that were derived through random forest classification of Sentinel-2, Landsat 8, and Sentinel-1 data at a resolution of 10 m. For each catchment, the percentage cover of 23 agricultural crop types distinguished by Blickensd¨orfer et al. (2022) was
quantified with ESRI ArcGIS Pro 2.9.0 and Spyder (Phyton 3.7). We added urban areas and forests for the year 2016 from the land use data of Griffiths et al. (2019) and processed it likewise. The time mismatch of crop maps (2017) and biological data (2010–2019) is likely partly compensated by the sheer size of the dataset (7748 sites). Moreover, results from analysis with land use data from 2018 differed only marginally (except for oilseeds) and would not change the general interpretation (Table S1). Although crops rotate between years for in- dividual fields, crop compositions tend to vary only slightly on larger spatial scales. However, farmers might have responded to extreme weather conditions such as excessive rainfall in 2018 (Blickensdorfer ¨ et al., 2022). Consequently, we chose to use ‘the year of biological data’
as a random factor in our models.
To decrease collinearity between predictors, avoid overfitting, and increase model stability, the 23 different crop types were grouped based on similarities in cultivation practices and assumed similarities in im- pacts on freshwater biota (Table S2), in particular fertilizer (Britz and Witzke, 2014) with highest application rates for maize and pesticide usage (Andert et al., 2015; Dachbrodt-Saaydeh et al., 2021). While crop- specific pesticide application rates are not available at a Germany-wide scale, Dachbrodt-Saaydeh et al. (2021) reported crop-specific pesticide treatment frequency indices, also referred to as the number of full doses (Sattler et al., 2007; Ferguson and Evans, 2010). Pesticide treatment frequencies for the different crop types ranged from 0.4 to 5.2 for her- bicides, 0 and 24.8 for fungicides, and 0 and 5.3 for insecticides in 2017 (Dachbrodt-Saaydeh et al., 2021). For permanent crops and vegetables, the highest pesticide treatment was reported, particularly concerning insecticides and fungicides, while the weakest pesticide applications were found for maize, on which exclusively herbicide application was reported (Dachbrodt-Saaydeh et al., 2021). The resulting land use groups were named “Intensive Cereals”, “Extensive Cereals”, “Oilseeds”,
“Permanent crops”, “Maize”, “Vegetables”, “Grassland”, “Urban” and
“Forest”, with mean percentage cover ranging between 1 and 35 % (Fig. S1). For general comparison of agriculture with urban areas, grassland, and forests, we combined all crop land use groups into a gross land use group “cropland”.
2.3. GLMM (generalized linear mixed model)
To quantify the associations between the land use groups and the ecological status, we fitted generalized linear mixed models (GLMMs) using the ‘gamlss’ package in r-studio (v5.2-0; Rigby and Stasinopoulos, 2005). The GLMMs were set up with a zero-one inflated beta regression (BEINF) and a logit link to decrease the dependency of the effect sizes on the gradient length (compare Mack et al., 2022). Prior to model fitting, we assessed collinearity with the variance inflation factor (VIF). Yielding values <2.5, we assumed small collinearity and kept all variables for the analysis. Also, only negligible spatial autocorrelation was observed (Figs. S2-S4), which was consequently not regarded in the models.
We built two models for each organism group. The general model contained the four gross land use groups (cropland, grassland, urban, and forest) as fixed variables. The specific model contained the nine more detailed land use groups (intensive cereals, extensive cereals, oil- seeds, permanent crops, maize, vegetables, grassland, urban, and forest) as fixed variables (Table S2). Additionally, four random variables were used in all models, namely “river type” following Pottgiesser and Som- merh¨auser (2008) (to account for differences in river size and altitude),
“year of biological sampling” (to account for variations in land use be- tween years), “federal state” (to account for possible differences in sampling), and “category” (indicating whether the sampling site was considered natural or ‘heavily modified’ in accordance with Article 4 of the EU Water Framework Directive). A visual check of the residual distribution against the predicted values and each variable yielded centered averages and symmetrical distributions. Finally, regression coefficients (log odds) including 95 % confidence intervals were calcu- lated using the r-package ‘parameters’ (v0.12.0; Lüdecke et al., 2020).
3. Results
3.1. Associations between river biota and cropland vs. other land use categories
Our first hypothesis (stronger negative associations between crop- land and river biota compared to urban land use) was largely supported by the results of the organism group-specific general models (Table S3).
For macroinvertebrates, the difference was most pronounced, with the regression coefficient for cropland (− 1.31) being >40 % larger compared to urban land use (− 0.92). Moreover, this difference can be considered reliable because the 95 % confidence intervals did not overlap, with − 0.73 to − 1.12 for urban land use and − 1.17 to − 1.45 for cropland. For macrophytes, regression coefficients differed only by 10
%: − 0.95 [− 1.19;− 0.71] for cropland and − 0.83 [− 1.20;− 0.45] for urban land use. For diatoms, the magnitude of land use associations were smaller, and although they differed by 40 %, the confidence in- tervals did overlap with regression coefficients of − 0.63 [− 0.73;− 0.53]
and − 0.44 [− 0.60;− 0.28] for cropland and urban land use, respectively.
Grassland and forest tended to have small negative or even positive regression coefficients (Table S2).
3.2. Associations between river biota and different crop type groups Hypothesis 2a (crop types differ in their associations with the ecological status) was partly supported by the specific models (Figs. 1–3). For macroinvertebrates, the regression coefficients of the six crop land use groups differed most strongly, ranging between log odds of
− 4.48 for permanent crops and 0.49 for extensive cereals, i.e. spanned a range of 4.97. Moreover, 95 % confidence intervals did not overlap, and differences were therefore reliable for the three groups of ‘permanent crops’ vs. ‘vegetables, maize and intensive cereals’ vs. ‘oilseeds and extensive cereals’ (Fig. 1). For macrophytes, the log odds only ranged between − 2.83 for permanent crops and − 0.48 for extensive cereals (range =2.35), and confidence intervals strongly overlapped for all crop type groups (Fig. 2). For diatoms, the range was even smaller, with regression coefficients from − 0.99 for maize to 1.05 for extensive ce- reals (range =2.04). However, confidence intervals of vegetables and extensive cereals did not overlap with ‘maize, intensive cereals, and oilseed’, indicating that at least these differences were reliable (Fig. 3).
The organism groups clearly differed in their sensitivity to the crop type groups, as predicted by hypothesis 2b (Figs. 1–3). For macro- invertebrates, the pesticide-intensive permanent crops and vegetables showed the strongest negative associations with regression coefficients of − 4.48 and − 1.82, respectively. The intensively fertilized crops ‘maize and intensive cereals’ had smaller negative regression coefficients of
− 1.33 and − 1.28, respectively, while oilseeds and extensive cereals had no clear associations. In contrast to the crops, grassland had a reliable positive association with a regression coefficient of 0.65 (Fig. 1). For macrophytes, a similar pattern was observed (Fig. 2). Pesticide-intensive permanent crops and vegetables had the strongest negative association with regression coefficients of − 2.83 and − 1.71 respectively, while intensive cereals associated with high nutrient application rates also had a relatively large negative regression coefficient of − 0.92. However, associations were smaller in magnitude compared to macro- invertebrates, and differences were less pronounced given that confi- dence intervals overlapped. For diatoms, a different pattern was observed (Fig. 3): Intensively fertilized crops like maize and intensive cereals, as well as oilseeds, showed the largest negative associations with regression coefficients of − 0.99, − 0.77 and − 0.75, respectively. In contrast, permanent crops and vegetables associated with high pesticide application rates showed no reliable relationship with confidence in- tervals overlapping the zero line. However, associations were smallest compared to macrophytes and macroinvertebrates, and most confidence intervals overlapped.
4. Discussion
The land use data recently published by Blickensd¨orfer et al. (2022) allowed to investigate the associations between specific crop types and different organism groups in much more detail compared to previous large-scale studies, which were restricted to the effect of cropland in general on river biota. The results clearly showed different associations between cropland land use at the catchment scale and the ecological status of macroinvertebrates, macrophytes and diatoms depending on the specific crop types. Given that these crop types typically differ in respect to application rates of fertilizers and pesticides, these results likely mirror the underlying effects of agrochemicals.
4.1. Associations between river biota and cropland vs. other land use categories
Our analysis underlines the differential associations between different types of catchment land use and river biota. The expectation of stronger associations of cropland compared to urban land use (hypoth- esis 1) was largely supported. This coincides with other recent studies suggesting agriculture to be the dominant stressor for river biota (Wolfram et al., 2021; Liess et al., 2021). Stressors characteristic for arable land (e.g. nutrients resulting from fertilizer use and pesticides) appear more severe on a nationwide scale, than typical stressors resulting from urban land use, such as sewage overflows, flash floods related to impervious land cover, and micropollutants (Weitere et al.,
2021) and urban pesticides (Gerecke et al., 2002) and nutrient stress (Müller et al., 2002) appear less relevant (Neumann et al., 2002).
Naturally, stressors originating from arable land and urban areas over- lap; e.g., parts of the agrochemicals found in wastewater treatment plants adjacent to urban areas originate from urban sources such as roads, railroads, and urban green spaces. However, the substantial pesticide contribution is likely to result from farm courtyard draining to urban point sources (Gerecke et al., 2002; Neumann et al., 2002). For- ests, the potentially natural vegetation in almost all parts of Germany, had an overall positive association with river biota, as also observed in other studies (e.g. Goss et al., 2020).
4.2. Associations between river biota and different crop type groups The crop type groups strongly differed in their associations with river biota (hypothesis 2a), likely caused by varying agricultural practices, in particular nutrient (Britz and Witzke, 2014) and pesticide application rates (Andert et al., 2015; Dachbrodt-Saaydeh et al., 2021). Pesticide application rates differ between permanent crops and extensive crops like maize by a factor of up to 15 (Dachbrodt-Saaydeh et al., 2021), while nutrient applications on maize exceed nutrient extensive crops such as permanent crops up to ten-fold (Britz and Witzke, 2014). These differences suggest stronger effects for pesticide- and nutrient-intensive crops compared to grassland (Blake et al., 2012) or crops such as rye and oat with much lower pesticide and fertilizer applications (Schulz et al., 2013; Mushtaq and Mehfuza, 2014). The relations between different Fig. 1.Zero one inflated GLMM with land use groups as fixed effects and the macroinvertebrate multimetric index as response (n =7748). Plot shows the regression coefficients of the different catchment land use groups (Intercept =0.00 [−0.12, 0.13]).
crop types and exerted stressors, suggested by the associations between land use and biota response, are further supported by preliminary analysis relating different crop types with pesticides monitored in rivers (unpublished results) and relating the different crop type groups with the saprobic index, a proxy for eutrophication and the subsequent biomass development (Table S4).
Also, as expected in hypothesis 2b, the organisms differ in their sensitivity to the crop type groups. For macroinvertebrates, the hierar- chical order of the association magnitudes between the crop type groups and biota response mirrors the pesticide application rates identified by Dachbrodt-Saaydeh et al. (2021) with strongest associations for per- manent crops and vegetables, for which highest insecticide and fungi- cide application rates were reported. These findings suggesting the importance of pesticides for macroinvertebrates are in line with findings of Liess et al. (2021). Nutrients seem less important, and fine sediments are likely less crop-type-dependent. Although fine sediments impair macroinvertebrates (Gieswein et al., 2019), their dynamics are strongly dependent on hydromorphological factors (Urbaniˇc, 2014) and stressor interactions (Piggott et al., 2012). Moderate nutrient concentrations may even benefit macroinvertebrates (Piggott et al., 2012), while effects turn negative as soon as oxygen depletion resulting from the decom- position of plant biomass is involved (Weitere et al., 2021). Nutrient concentrations might be too low in most German streams to cause these effects. Macrophytes also appear to respond most strongly to pesticide- intensive crops, coinciding with mesocosm studies finding strong
effects of agricultural herbicides (e.g. Ribeiro et al., 2019), while O’Hare et al. (2018) contrastingly suggested strong nutrient sensitivity. How- ever, large confidence intervals hinder interpretation, and other factors such as river morphology (Kaijser et al., 2022) or river management (Bączyk et al., 2018) may be more important for macrophyte occurrence.
For diatoms, the associations likely mainly relate to nutrient applica- tions, with the strongest associations for the most nutrient-intensive crop type groups, maize and intensive cereals (Britz and Witzke, 2014) and weaker associations for the insecticide- and fungicide- intensive permanent crops and vegetables. Although the overall pesti- cide application rates are lower for the crop types with strongest diatom response associations (maize, intensive cereals, oilseeds), the share of herbicides applied is higher (Andert et al., 2015; Dachbrodt-Saaydeh et al., 2021). These findings highlight the importance of eutrophication effects for diatoms and match the literature (O’Hare et al., 2018), accompanied by potential effects of herbicides (Debenest et al., 2010).
Other than for macroinvertebrates and macrophytes, grassland showed a negative association with diatoms, as most grassland is fertilized in addition to nutrients from livestock (Mouri and Aisaki, 2015). Overall, the crop-type specific differences were most pronounced for macro- invertebrates, while macrophytes and diatoms showed large confidence intervals or smaller regression coefficients, respectively. This coincides with findings from Schürings et al. (2022) suggesting macro- invertebrates are most strongly affected by agriculture. Though we did not investigate causal relationships, our results provide support for the Fig. 2. Zero one inflated GLMM with land use groups as fixed effects and the macrophyte multimetric index as response (n =2905). Plot shows the regression coefficients of the different catchment land use groups (Intercept = −0.16 [−0.36, 0.05]).
detrimental effects of pesticide-intensive crops.
To prevent misinterpretation of the positive regression coefficients of extensive cereals and grasslands, which are at a similar magnitude as forests, it is important to consider that all regression coefficients in one model are interdependent. Large shares of extensive cereals and grass- lands in a catchment cause lower percentages of more harmful land use types. Moreover, grassland incorporates various types that we could not discriminate: pasture with livestock effects (Mouri and Aisaki, 2015) and (natural) grasslands (Blickensd¨orfer et al., 2022). Extensive cereals such as rye and oat grow on poor and acid soil and are, besides being main crops, also used as intermediate cover crops, facilitating nutrient retention and reducing soil erosion. They are also known for their allelochemical properties, suppressing weeds, so fewer herbicides and plowing of land are needed (Schulz et al., 2013; Gataneh et al., 2021).
5. Conclusion
The results suggest against using oversimplified approaches to ac- count for agricultural effects such as the sheer cover of arable land and grassland as a proxy for agricultural stress. This study indicates strong differences in agricultural associations on river biota, mainly depending on crop types, which should be accounted for when assessing land use stress. Using multiple organism groups best reflects agricultural impacts.
The strong associations of pesticide-intensive crop types call for the implementation of buffer strips to reduce pesticide inputs and, ideally,
the implementation of more pesticide-free, environmentally friendly farming practices such as organic agriculture or permaculture to improve the ecological status of rivers.
CRediT authorship contribution statement
Christian Schürings: Conceptualization, Methodology, Writing- Original draft preparation, Writing- Reviewing and Editing.
Jochem Kail: Conceptualization, Methodology, Writing- Rewriting and Editing, Supervision.
Willem Kaijser: Methodology, Writing- Reviewing and Editing.
Daniel Hering: Conceptualization, Methodology, Writing- Review- ing and Editing, Supervision.
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Data availability
Data and code for reproducing numbers and figures shown are available at https://github.com/cs7792b/Effects_of_crop_types Fig. 3. Zero one inflated GLMM with land use groups as fixed effects and the diatom multimetric index as response (n =3402). Plot shows the regression coefficients of the different catchment land use groups (Intercept = − 1.25 [−1.34, −1.17]).
Acknowledgements
This work was financially supported by a scholarship funding from the German Federal Environmental Foundation (DBU) and by the proj- ect AQUATAG funded by the German Ministry of Education and Research (BMBF), project number 02WRM046, which is gratefully acknowledged. We are grateful to the German federal environmental departments, who provided the biological data and to Lukas Blick- ensd¨orfer et al. for the availability of the land use maps.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://doi.
org/10.1016/j.scitotenv.2023.168825.
References
Andert, S., Bürger, J., Gerowitt, B., 2015. On-farm pesticide use in four Northern German regions as influenced by farm and production conditions. Crop Prot. 75, 1–10.
https://doi.org/10.1016/j.cropro.2015.05.002.
Artigas, J., Majerholc, J., Foulquier, A., Margoum, C., Volat, B., Neyra, M., Pesce, S., 2012. Effects of the fungicide tebuconazole on microbial capacities for litter breakdown in streams. Aquat. Toxicol. 122, 197–205. https://doi.org/10.1016/j.
aquatox.2012.06.011.
Bączyk, A., Wagner, M., Okruszko, T., Grygoruk, M., 2018. Influence of technical maintenance measures on ecological status of agricultural lowland rivers–systematic review and implications for river management. Sci. Total Environ. 627, 189–199.
https://doi.org/10.1016/j.scitotenv.2018.01.235.
Blake, W.H., Ficken, K.J., Taylor, P., Russell, M.A., Walling, D.E., 2012. Tracing crop- specific sediment sources in agricultural catchments. Geomorphology 139, 322–329.
https://doi.org/10.1016/j.geomorph.2011.10.036.
Blickensd¨orfer, L., Schwieder, M., Pflugmacher, D., Nendel, C., Erasmi, S., Hostert, P., 2022. Mapping of crop types and crop sequences with combined time series of Sentinel-1, Sentinel-2 and Landsat 8 data for Germany. Remote Sens. Environ. 269, 112831 https://doi.org/10.1016/j.rse.2021.112831.
B¨ohmer, J., Rawer-Jost, C., Zenker, A., Meier, C., Feld, C.K., Biss, R., Hering, D., 2004.
Assessing streams in Germany with benthic invertebrates: development of a multimetric invertebrate based assessment system. Limnologica 34 (4), 416–432.
https://doi.org/10.1016/S0075-9511(04)80010-0.
Britz, W., Witzke, P., 2014. CAPRI Model Documentation 2014. Available online at. htt p://www.capri-model.org/docs/capri_documentation.pdf. Viewed 02 March 2023.
CBD, 2022. Kunming-Montreal Global Biodiversity Framework. UN Environment Programme, Report No. CBD/COP/15/L.25. https://www.cbd.int/doc/c/e6d3/
cd1d/daf663719a03902a9b116c34/cop-15-l-25-en.pdf. Viewed 02 March 2023.
Dachbrodt-Saaydeh, S., Sellmann, J., Strassemeyer, J., Schwarz, J., Klocke, B., Krengel, S., Kehlenbeck, H., 2021. Netz Vergleichsbetriebe Pflanzenschutz Jahresbericht 2017 Analyse der Ergebnisse der Jahre 2007-2017, Julius Kühn- Institut, Report No. 210. https://www.nap-pflanzenschutz.de/fileadmin/SITE_MAS TER/content/Bilder/Indikatoren_Forschung/Erfassung_realer_PSM-Anwendungen /JKI-210_Vergleichsbetriebe_2017.pdf. Viewed 02 March 2023.
Davis, N.G., Hodson, R., Matthaei, C.D., 2022. Long-term variability in deposited fine sediment and macroinvertebrate communities across different land-use intensities in a regional set of New Zealand rivers. NZ J. Mar. Freshwat. Res. 56 (2), 191–212.
https://doi.org/10.1080/00288330.2021.1884097.
Debenest, T., Silvestre, J., Coste, M., Pinelli, E., 2010. Effects of pesticides on freshwater diatoms. Rev. Environ. Contam. Toxicol. 87-103 https://doi.org/10.1007/978-1- 4419-1352-4_2.
Ferguson, A., Evans, N., 2010. Reducing pesticide inputs in winter cropping systems in the UK. In: WCCS, Case Study, Guide Number 3, 1–8. http://www.endure-network.
eu/content/download/6040/45288/file/Winter%20Crops%20%20Based%20Cro pping%20Systems%20Case%20Study%20Guide%20Number%203.pdf. Viewed 19 October 2023.
Foley, J.A., DeFries, R., Asner, G.P., Barford, C., Bonan, G., Carpenter, S.R., Chapin, F.S., Coe, M.T., Daily, G.C., Gibbs, H.K., Helkowski, J.H., Holloway, T., Howard, E.A., Kucharik, C.J., Monfreda, C., Patz, J.A., Prentice, I.C., Ramankutty, N., Snyder, P.K., 2005. Global consequences of land use. Science 309, 570–574. https://doi.org/
10.1126/science.1111772.
Gataneh, F.A., Forsido, S.F., Yetenayet, B.T., Addisu, A.A., Minbale, A.T., Endale, A., 2021. Traditional food processing practices of oats (Avena sativa) and its contribution to food security in Gozamin district of northwest Ethiopia. Afr. J. Food Agric. Nutr. Dev. 21 (5), 18083–18100. https://doi.org/10.18697/
ajfand.100.19810.
Gerecke, A.C., Sch¨arer, M., Singer, H.P., Müller, S.R., Schwarzenbach, R.P., S¨agesser, M., Popow, G., 2002. Sources of pesticides in surface waters in Switzerland: pesticide load through waste water treatment plants––current situation and reduction potential. Chemosphere 48 (3), 307–315. https://doi.org/10.1016/S0045-6535(02) 00080-2.
Gieswein, A., Hering, D., Feld, C.K., 2017. Additive effects prevail: the response of biota to multiple stressors in an intensively monitored watershed. Sci. Total Environ. 593, 27–35. https://doi.org/10.1016/j.scitotenv.2017.03.116.
Gieswein, A., Hering, D., Lorenz, A.W., 2019. Development and validation of a macroinvertebrate-based biomonitoring tool to assess fine sediment impact in small mountain streams. Sci. Total Environ. 652, 1290–1301. https://doi.org/10.1016/j.
scitotenv.2018.10.180.
Goss, C.W., Sullivan, S.M.P., Goebel, P.C., 2020. Effects of land-cover transitions on emerging aquatic insects and environmental characteristics of headwater streams in an agricultural catchment. River Res. Appl. 36 (7), 1097–1108. https://doi.org/
10.1002/rra.3639.
Griffiths, P., Nendel, C., Hostert, P., 2019. Intra-annual reflectance composites from Sentinel-2 and Landsat for national-scale crop and land cover mapping. Remote Sens.
Environ. 220, 135–151. https://doi.org/10.1016/j.rse.2018.10.031.
Haase, P., Lohse, S., Pauls, S., Schindehütte, K., Sundermann, A., Rolauffs, P., Hering, D., 2004. Assessing streams in Germany with benthic invertebrates: development of a practical standardised protocol for macroinvertebrate sampling and sorting.
Limnologica 34 (4), 349–365. https://doi.org/10.1016/S0075-9511(04)80005-7.
Hahn, M., Schotth¨ofer, A., Schmitz, J., Franke, L.A., Brühl, C.A., 2015. The effects of agrochemicals on Lepidoptera, with a focus on moths, and their pollination service in field margin habitats. Agric. Ecosyst. Environ. 207, 153–162. https://doi.org/
10.1016/j.agee.2015.04.002.
Hering, D., Feld, C.K., Moog, O., Ofenb¨ock, T., 2006. Cook book for the development of a Multimetric Index for biological condition of aquatic ecosystems: Experiences from the European AQEM and STAR projects and related initiatives. In: The Ecological Status of European Rivers: Evaluation and Intercalibration of Assessment Methods.
Springer, Dordrecht, pp. 311–324. https://doi.org/10.1007/978-1-4020-5493-8_22.
Joppa, L.N., O’Connor, B., Visconti, P., Smith, C., Geldmann, J., Hoffmann, M., Burgess, N.D., 2016. Filling in biodiversity threat gaps. Science 352 (6284), 416–418. https://doi.org/10.1126/science.aaf3565.
Jwaideh, M.A., Sutanudjaja, E.H., Dalin, C., 2022. Global impacts of nitrogen and phosphorus fertiliser use for major crops on aquatic biodiversity. Int. J. Life Cycle Assess. 27 (8), 1058–1080. https://doi.org/10.1007/s11367-022-02078-1.
Kaijser, W., Hering, D., Lorenz, A.W., 2022. Reach hydromorphology: a crucial environmental variable for the occurrence of riverine macrophytes. Hydrobiologia 849 (19), 4273–4285. https://doi.org/10.1007/s10750-022-04983-w.
Liess, M., Liebmann, L., Vormeier, P., Weisner, O., Altenburger, R., Borchardt, D., Brack, W., Chatzinotas, A., Escher, B., Foit, K., Gunold, R., Henz, S., Hitzfeld, K.L., Schmitt-Jansen, M., Kamjunke, N., Kaske, O., Knillmann, S., Krauss, M., Küster, E., Link, M., Lück, M., Moder, M., Müller, A., Paschke, A., Sch¨ ¨afer, R.B.,
Schneeweiss, A., Schreiner, A.C., Schulze, T., Schüürmann, G., Tümpling, W.V., Weitere, M., Wogram, J., Reemtsma, T., 2021. Pesticides are the dominant stressors for vulnerable insects in lowland streams. Water Res. 201, 117262 https://doi.org/
10.1016/j.watres.2021.117262.
Lorente, C., Causape, J., Glud, R.N., Hancke, K., Merch´an, D., Mu˜niz, S., Navarro, E., 2015. Impacts of agricultural irrigation on nearby freshwater ecosystems: the seasonal influence of triazine herbicides in benthic algal communities. Sci. Total Environ. 503, 151–158. https://doi.org/10.1016/j.scitotenv.2014.06.108.
Lüdecke, D., Ben-Shachar, M., Patil, I., Makowski, D., 2020. Extracting, computing and exploring the parameters of statistical models using R. J. Open Source Softw. 5 (53), 2445. https://doi.org/10.21105/joss.02445.
Mack, L., de la Hoz, C.F., Penk, M., Piggott, J., Crowe, T., Hering, D., Birk, S., 2022.
Perceived multiple stressor effects depend on sample size and stressor gradient length. Water Res. 226, 119260 https://doi.org/10.1016/j.watres.2022.119260.
Mouri, G., Aisaki, N., 2015. Using land-use management policies to reduce the environmental impacts of livestock farming. Ecol. Complex. 22, 169–177. https://
doi.org/10.1016/j.ecocom.2015.03.003.
Müller, K., Bach, M., Hartmann, H., Spiteller, M., Frede, H.G., 2002. Point-and nonpoint- source pesticide contamination in the Zwester Ohm catchment, Germany. J. Environ.
Qual. 31 (1), 309–318. https://doi.org/10.2134/jeq2002.3090.
Mushtaq, A., Mehfuza, H., 2014. A review on oat (Avena sativa L.) as a dual-purpose crop. Sci. Res. Essays 9 (4), 52–59. https://doi.org/10.5897/SRE2014.5820.
Neumann, M., Schulz, R., Sch¨afer, K., Müller, W., Mannheller, W., Liess, M., 2002. The significance of entry routes as point and non-point sources of pesticides in small streams. Water Res. 36 (4), 835–842. https://doi.org/10.1016/S0043-1354(01) 00310-4.
O’Hare, M.T., Baattrup-Pedersen, A., Baumgarte, I., Freeman, A., Gunn, I.D., L´az´ar, A.N., Sinclair, R., Wade, A.J., Bowes, M.J., 2018. Responses of aquatic plants to eutrophication in rivers: a revised conceptual model. Front. Plant Sci. 9, 451.
https://doi.org/10.3389/fpls.2018.00451.
Piggott, J.J., Lange, K., Townsend, C.R., Matthaei, C.D., 2012. Multiple stressors in agricultural streams: a mesocosm study of interactions among raised water temperature, sediment addition and nutrient enrichment. PLoS One 7 (11), e49873.
https://doi.org/10.1371/journal.pone.0049873.
Pottgiesser, T., Sommerh¨auser, M., 2008. Erste Überarbeitung der Steckbriefe der deut- schen Fließgewassertypen. http://www.wasserblick.net/servlet/is/18727/. Viewed ¨ 02 March 2023.
Ribeiro, V.H.V., Alencar, B.T.B., Dos Santos, N.M.C., da Costa, V.A.M., Dos Santos, J.B., Francino, D.M.T., Silva, D.V., 2019. Sensitivity of the macrophytes Pistia stratiotes and Eichhornia crassipes to hexazinone and dissipation of this pesticide in aquatic ecosystems. Ecotoxicol. Environ. Saf. 168, 177–183. https://doi.org/10.1016/j.
ecoenv.2018.10.021.
Rigby, R.A., Stasinopoulos, D.M., 2005. Generalized additive models for location, scale and shape, (with discussion). Appl. Stat. 54, 507–554. https://doi.org/10.1111/
j.1467-9876.2005.00510.x.
Sattler, C., K¨achele, H., Verch, G., 2007. Assessing the intensity of pesticide use in agriculture. Agric. Ecosyst. Environ. 119 (3–4), 299–304. https://doi.org/10.1016/j.
agee.2006.07.017.
Schaumburg, J., Schranz, C., Stelzer, D., Vogel, A., Gutowski, A., 2012.
Verfahrensanleitung für ¨okologische Bewertung von Fließgew¨assern zur Umsetzung der EU-Wasserrahmenrichtlinie: Makrophyten und Phytobenthos. Bayrisches Landesamt für Umwelt. https://www.flussgebiete.nrw.de/system/files/atoms/files/
d9_kap-121-phylib-fg.pdf.
Schulz, R., Liess, M., 1999. A field study of the effects of agriculturally derived insecticide input on stream macroinvertebrate dynamics. Aquat. Toxicol. 46 (3–4), 155–176.
https://doi.org/10.1016/S0166-445X(99)00002-8.
Schulz, M., Marocco, A., Tabaglio, V., Macias, F.A., Molinillo, J.M., 2013.
Benzoxazinoids in rye allelopathy-from discovery to application in sustainable weed control and organic farming. J. Chem. Ecol. 39 (2), 154–174. https://doi.org/
10.1007/s10886-013-0235-x.
Schürings, C., Feld, C.K., Kail, J., Hering, D., 2022. Effects of agricultural land use on river biota: a meta-analysis. Environ. Sci. 34, 124. https://doi.org/10.1186/s12302- 022-00706-z.
Secchi, S., Gassman, P.W., Jha, M., Kurkalova, L., Kling, C.L., 2011. Potential water quality changes due to corn expansion in the Upper Mississippi River Basin. Ecol.
Appl. 21 (4), 1068–1084. https://doi.org/10.1890/09-0619.1.
Stehle, S., Schulz, R., 2015. Agricultural insecticides threaten surface waters at the global scale. Proc. Natl. Acad. Sci. 112 (18), 5750–5755. https://doi.org/10.1073/
pnas.1500232112.
UBA, 2022. Die Wasserrahmenrichtlinie Gew¨asser in Deutschland 2021 Fortschritte und Herausforderungen. German Environmental Agency. https://www.umweltbundesa
mt.de/sites/default/files/medien/1410/publikationen/221010_uba_fb_wasserrich tlinie_bf.pdf. Viewed 02 March 2023.
Urbaniˇc, G., 2014. Hydromorphological degradation impact on benthic invertebrates in large rivers in Slovenia. Hydrobiologia 729 (1), 191–207. https://doi.org/10.1007/
s10750-012-1430-4.
Wasson, J.G., Villeneuve, B., Iital, A., Murray-Bligh, J., Dobiasova, M., Bacikova, S., Timm, H., Pella, H., Mengin, N., Chandesris, A., 2010. Large-scale relationships between basin and riparian land cover and the ecological status of European rivers.
Freshw. Biol. 55, 1465–1482. https://doi.org/10.1111/j.1365-2427.2010.02443.x.
Weitere, M., Altenburger, R., Anlanger, C., Baborowski, M., B¨arlund, I., Beckers, L.M., Brauns, M., 2021. Disentangling multiple chemical and non-chemical stressors in a lotic ecosystem using a longitudinal approach. Sci. Total Environ. 769, 144324 https://doi.org/10.1016/j.scitotenv.2020.144324.
Wolfram, J., Stehle, S., Bub, S., Petschick, L.L., Schulz, R., 2021. Water quality and ecological risks in European surface waters–monitoring improves while water quality decreases. Environ. Int. 152, 106479 https://doi.org/10.1016/j.
envint.2021.106479.
Zhou, M., Wu, S., Zhang, Z., Aihemaiti, Y., Yang, L., Shao, Y., Chen, Z., Jiang, Y., Jin, C., Zheng, G., 2022. Dilution or enrichment: the effects of flood on pollutants in urban rivers. Environ. Sci. Eur. 34 (1), 1–12. https://doi.org/10.1186/s12302-022-00639- 7.